Exploring Total Border

[NHhighlander]

In the words of Scotland's foremost tartan historian, Peter Eslea MacDonald, a total border pattern uses the
selvedge pattern around the whole plaid.
There was a time when I had no idea what that meant; therefore, in the interest to help someone exploring this matter, a selvedge is the "natural" edge of the cloth, "self-edge," also "selvage," where the threads that constitute the weft end their trip through the warp, and go back. A selvedge pattern makes that border part of the cloth distinctive, by a choice of colors different from the rest, or by the pattern of weaving, different from the 2/2 twill.

\\\\\\\\
\\\\\\\\ 2/2 twill (tartan standard surface pattern)
The standard is that tartans are woven with twill in one single direction, a diagonal going from top left, down to lower right at strict 45 degrees.
weave0101twill.jpg

////\\\\///\\\
////\\\\///\\\ herringbone
Happens when a segment is twill in one direction, the next in the opposite, then flip-flop again as many times as desired. I did here 10+1 threads per direction.
For the ends, top and bottom, it's done by inverting the order or sequence of lifting the shafts. For the sides, right or left, it's built-in by the thread sequence in setting up the heddles.
weave0101herringbone.jpg

\\\///\\\///
///\\\///\\\ birdseye
\\\///\\\///
Happens when the sequence of lifting the shafts is inverted, and the heddles were set for herringbone.
weave0101birdseye.jpg

Oh, it's hopeless to explain every word... Seems that for each, I need to pull at least two more, without counting that I am far from being an expert in weaving nomenclature (corrections welcome!). Sigh. Let's move on.
Funny how weaving is the gift that keeps on giving in that sense. Browsing a new book I received today, parts read like the Jabberwocky.
Yay!

A lot of things will make better sense if you read or at least browse MacDonald's papers on these matters; that is how I started. He writes quite clearly, for such a complex topic.

First, I was a little intrigued. Then, fascinated.
Until a couple of weeks ago, when I decided to start hands-on experimenting. And studying and learning for real. And joined this forum.

Here, for a good start:

• https://www.scottishtartans.co.uk/Tw...h%20County.pdf
• https://www.scottishtartans.co.uk/Tw...ish_County.pdf
• https://www.scottishtartans.co.uk/Tr...e_patterns.pdf

Last week I completed a "theoretic" solution to the total border, using strips of paper, and then a computer graphics program. Then purchased a table jack loom on Craigslist. (Yes, I want to build my own, but my wife wisely pointed out the time delay, as she saw me fretting to start weaving...)

Why it matters

It's cool.
It's unique.
It's complicated.
The aesthetics are logic.

That kind of making adventure is to me like a candle to a moth...


Many lessons learned. There is hope

1. I should not leave my day job.
2. Bands of color, especially a selvedge mark (a distinctive color band running along the selvedge) should match/overlap in the axis of the mirror on the herringbone/birdseye.
   Best if it takes the whole width of the ///\\\ mirror pattern. That wide white looks pretty decent. While a narrow mirror pattern is centered on the green, I wasn't symmetrical with the width of the dark red, the end result for those sectors looks wrong.
   weave0101.jpg
   
3. make plenty shuttles, the first one I made is working fine. (mine is on the top right, center is a real Leclerc. Big plastic straws for bobbins)
   shuttles.jpg
4. The right tools make a difference. This loom is NOT able, with someone as clumsy as I, to do quality work.
   Fine loom for initial learning.
   theloom.jpg
   Jumps in tension is bad for you, it's no good for weaving, either.
   
   • Transform the jack loom into a countermarch loom. Sorry, I won't explain that, I'll just do it.
   • A friction brake for the warp beam.
   • A worm-gear device for the cloth beam
5. This is the first week in my life that I am using a loom with heddles. Skills in more uniformly running the shuttles, etc., hopefully will come with more practice.
6. 45-degrees, square squares, must focus on.
7. Careful sewing line-to-line looks SO much better than "just sewing." (probably will redo. I started in the center, and sort of made some sense by the end of what is on the left side of the picture)

Setting up the threads
The hardest challenge so far was to figure out the "logic" for weaving birdseye with 4 heddles. It has been done by others, so, it can be done. Took me two weeks of doodles while waiting here or there, then 4 hours with cut pieces of paper on the kitchen counter. Not trivial. But so much fun!

What I love of this, is that it's all geometry. Pure maths. And logic.
To the point that, at some moment I was confused, couldn't figure something; I decided to just trust the principles of symmetry, et voilà, it worked!
It's all binary; that is, a thread is up, or it's down.

Each heddle carries one thread, and one thread only.
Heddles go in shafts. "Canonically" we use 4 shafts for tartan weaving.
Each shaft carries 1/4 of the threads that form the weft.

The pattern for each shaft is usually something like this:
1000
that being, one thread is caught in a heddle, 3 threads pass by. That would be shaft a, a sequence of
1000100010001000... etc.

The next shaft, b, will catch one of the threads that was not captured by a. Thusly
01000100... etc

shaft c will go like
00100010... etc.


The basic pattern to weave 2/2 twill is 2 lines up, 2 lines down, for example, 1100, repeated.
1100110011001100... etc.

We will need to lift shaft a and shaft b at the same time. In my case, with a table jack loom, that is done with a lever A that has two strings, one going to each shaft.

then, the next line is the same principle of two-up and two down, but moving one line to the right. In binary, divide by 10, get
011001100110011001
Nobody knows binary? no problem!
just tie together shaft b and shaft c, to lever B

Figure out lever C by yourself, and, let me help, for lever D[/B] you tie together shaft d plus shaf a.

The sequence line by line is repeats of those lines.
Each line (that being, each thread of the weft), is determined by lifting a pair of shafts using the corresponding lever. The tunnel-like opening in the warp that is made that way as threads of the warp form two planes is called the shed, and we run the shuttle through that "tunnel.'

1100 pull lever A (pass the shuttle with one line of weft)

0110 lever B (pass the shuttle the other way)
0011 etc.
1001

then, if you continue doing the same, A B C D followed by A B C Dover and over, you got twill
1100
0110
0011
1001
1100
0110

BUT, if you inverse your sequence, as in A B C D followed by C B A D, then A B C D again, you get herringbone
\\\
///
\\\

1100 A
0110 B
0011 C
1001 D
0011 C
0110 B
1100 A
0110 B
0011 C

Now, and here is where the fun gets fun, think about mesing with those heddles back when you were setting up your warp.
01000001000 100010001000... shaft a, normal, as if a twill
00100010100 010001000100 notice that shaft b is getting funny...
00010100010 001000100010 and then c is hilarious
10001000001 000100010001 while d is evidently ridiculous

Notice the space in each line of the description. It's just to help see how the pattern before that space is forming chevrons (that is, herringbone on its head) while the rest is just plain twill.

Tah-dah! When attached to the levers, as you go ABCDABCDABCD, you get herringbone/chevron as selvedge pattern, followed by 2/2 twill for the rest.

Then, and brace yourselves, if you go ABCDCBDABCDCBDA, BirdsEye forms on the edge! And herringbone in the rest, exactly what you want to begin and end (or in my case, the center also).

Pfew!

Alas!, this is turning into a (bad) treatise of maths-in-arts... And I haven't even gotten to the logic for the shafts in a loom to be paired, how those pairs behave like DNA nitrogenous bases in base pairs, the famous "helix." That is, shaft a can pair with either b or d, joined as levers A or D respectively, but never with c. It would be possible to weave a genome using the 2/2 twill principles... I'm sure it's been done already, it's so obvious, so no need for me to dwell on it.

This is, more precisely, what I did, for each shaft:
BaselineTotalBorder.jpg
Then, this is what I expected to get (and got!)
totalBorder-20.jpg

playing with different sizes for the birdseye (I used the pattern that was 20 lines wide)
BaseChevroneBirdsEye.jpg


parts.jpg

[figheadair]

If you want a real challenge, try birdseye across the whole a the pattern. It requires the weaver to maintain two sets of counting concurrently, sett and treadle-pattern. It's extraordinarily complex and not something that I recommend with others around but the result is stunning for a decorative piece.

481476548_1150416836874691_3076817068169463806_n.jpg

[NHhighlander]

If you want a real challenge, try birdseye across the whole a the pattern. It requires the weaver to maintain two sets of counting concurrently, sett and treadle-pattern. It's extraordinarily complex and not something that I recommend with others around but the result is stunning for a decorative piece.

Wow!
Indeed.

I've been munching also on birdseye of different widths, maybe even lengths, so as to achieve an even distribution of geometric elements as the width of the color columns in the sett differ, as they generally do.

Off my head, I'd say that these weaving patterns set in the heddles (probably) must share a meaningful common denominator, as in a 20-wide birdseye/10-thread herringbone band would fit nicely next to a 4-wide or 10-wide birdseye / 2 or 5-herringbone, but not so easy neighboring a 6-wide. Meaning, many "canonical" sett patterns would suffer from this exercise, right now totally hypothetical anyway.
However, empowered by my extensive lifetime experience of 2 pieces of fabric woven, I have so far also learned that what looks OK in the cartoon doesn't necessarily translate so well to the loom, and vice versa. Example, the "fail" with the green/red/darkRed patterns, where it turns out that it is actually essential to make the mirror image be fully symmetrical for a good effect, and the unexpected "win" of the strong contrast of white on top of the other colors - which would have benefited from one more thread each on bottom and top to complete the pointy corner.

I find interesting the choice made by the weaver of the Maclaine of Lochbuie plaid, when facing the challenge of that pesky 8-thread green band. Preemptively pulling the excess 2 threads into the bands closer to the edge, when deeper into the body of the plaid he obviously was very careful to fit the 5 10-wide herringbone so it fits whole within the 50-wide color bands. That obviously was "planned", as the choice has to be made when setting up the heddles.
I'm just starting to dig into that paper, it seems to cover several of my doubts. I had browsed briefly before, it being my inspiration for my experiments with herringbone/chevron sides and birdseye corner.

Very much in agreement, counting treadle-pattern to conform to the sett, it's... complicated, famous understatement.
I followed this cartoon, cut and glued together the segments to form a long tape, and only once had to unravel 1.5 inches .
CartoonChevronBirdsEye.jpg

My kid sister advices using an abacus... I am growing fond to think that the view of the pattern as it builds up is at the very least a sanity check, yet perhaps the primary indicator of what should be going on and what is the next step. I'll still keep to being guided by the cartoon for at least a while probably. However, I have a feeling that "true expert craftsmen" back then were more into the intuitive/visual, achieved through experience.

And a sigh! note:
Re-reading this morning the traditional selvedge patterns paper, I noticed that some definitions that I put in my original post are wrong. I'll have to rewrite the whole thin; it certainly doesn't read easy, either.

[jsrnephdoc]

I consider myself lucky to have "advanced" to being able to distinguish pleating to the sett from to the stripe/military, and from box pleating, and confessing that I'm totally at the mercy of whomever is sewing my kilt to advise me which i should choose for any given tartan.

However, I DID learn, while touring the Lochcarraon Mill in summer 2023 that the need for skill in tartan production is no less important than the need for patience. Near the end of our tour we were shown the "Oops" room (I don't think they called it that), in which a VERY friendly matron was happily repairing a freshly woven (I think 30 meter) bolt of cloth that suffered from a single-thread repeating error. My eyes would have requested a week off from work by the time she finished a single iteration of her task!

[NHhighlander]

/snip/
the need for skill in tartan production is no less important than the need for patience.

When I was pulling this piece of frabric off the loom, I felt rather embarassed in particular about the edge. Because of the impossibility to keep an even pressure in this loom, already did I expect the diagonal symmetry to be off, and I mentioned already the mistakes I had made regarding the color scheme.

The edge, that I thought that should have been able to control better.
OK, sure, it's just the second ever piece of fabric I made in a loom with heddles, but I have woven some in my early days, and my edges were "better," tighter.
Let's have this one here for public display:

weave0101border.jpg

Then, as I was playing with it, and sewing the two halves together, that border and the inherent "ruggedness" of noob work kind of grew on me.
It does help that, generally speaking, I am a fan of naif art, and "wabi sabi" is something that endears any object to me.

Yeeees. I should, and will, strive for quality, the "arete" of Pirsig's "Zen and..."
Squares of the sett MUST be square, to a reasonable expression of square, and certainly the color patterns must match any funny thing of herringbone, etc., I come up with in the selvedge area, the twill must be very close to 45 degrees. The edge will be more even, just because it's a function of the motion of the shuttle, which will become ingrained as muscle memory.
And, funny enough, I've seen this kind of rough edges in some pictures of handmade tartan, and lines that are not completelly straight, etc. If ever somewhere someone looks at my kilt with a hard face, and says, "it looks like you made this yourself!", I'll totally take that as a compliment...

Patience is assuredly a most important ingredient, I agree with you.
Right now I'm in "honeymoon mode." Excited because it is all new.
It remains to be seen if I can keep things moving, as other priorities come up, etc., and routine needs borne on patience.

Near the end of our tour we were shown the "Oops" room (I don't think they called it that), in which a VERY friendly matron was happily repairing a freshly woven (I think 30 meter) bolt of cloth that suffered from a single-thread repeating error. My eyes would have requested a week off from work by the time she finished a single iteration of her task!

That was my dad. Or my sister.
If I have to face that, probably I'll just pay the fee to register whatever came out as a new tartan pattern LOL.
Yeah, they probably run no risk of me taking business away from them!

Edit: apologies, I see your point about skill. Agree. We'll build it, I hope, even though lack of patiece is what I tend to fear most in life.

[NHhighlander]

not "exactly" related to total border, tools are anyway part of moving forward the project.
Very kindly, the lady that sold me the loom included one shuttle, a Leclerc, light and smooth. Tartans have multiple colors, ergo, multiple shuttles is something that adds to the pleasure and productivity. I immediatley made one, pictured in the original post next to the loom, but for moving forward, I'll need a few more.

Early on, I wisely gave up on any pretense to carve out the shuttle out of a single piece of wood. Then, my attempt to turn my cheapo bench drill press into a functional wood lathe was "promising," but not.
I'm happy of what I eventually came up with: two slats of wood, glued to two blocks, then sanded away, I do have a nice disc sander.
Starting a bit earlier than the 2:57 pm timestamp.
Raw blanks:
shuttles1.jpg

"squared" from the rough, 3:22 pm
shuttles2.jpg

Rounded to "final" shape, 4:14pm
shuttles3.jpg

4:56pm, sanded to 180 and then 400 grit, done for today (Friday)
shuttles4.jpg

A good two hours so far. Still need the notches and the athingamabob stick for the bobbin.

Is it worth it? Well, if I actually were paid what I'm worth per hour, then, no.

Plastic Leclerc go for $22 a pop http://www.leclerclooms.com/cat2014a.htm , then what the lady gave me would set me back only $51 (nice unexpected add-on to my purchase, much grateful) (plus shipping and tariffs, eh!) But then, the pleasure of being to some extent self-sufficient, priceless! (I know that's fictional. I had to Amazon-in the sanding paper discs, that was $18, etc)

But I'm not paid much or very often, then, yes. And it's fun!

My collection, on display,
shuttles5.jpg

the one on the foreground was a special treat. In the evening, I grabbed one last blank that I had set aside, and the Swiss-army knife. This one probably took a bit more time, and almost made a blister... But ends up looking good, a final pass of 180 and then 400-grit to round the edges of the chip carving.
In the back are my Leclerc and the prototype, which actually was used for real already.
Sharp eyes will notice that the one in the center already has the notches for the bobbin axle, an easier than expected task with the bench drill press.